Nitrogen deficiency is a critical factor affecting tea cultivation, leading to significant reductions in yield and quality. Its detection at early stages is of significant importance. This study aims to spot nitrogen shortage in tea leaves early by using Gaussian Blur to reduce noise and improve color features. The model identifies early signs of stress, like yellowing, which point to a lack of nitrogen. The color data is then analyzed using Random Forest, Decision Tree, Naive Bayes, XGBoost, and Extreme XGBoost. Among these, Random Forest performed the best with an accuracy of 86%, showing it can handle complex data well. To make the results easier to understand, LIME (Local Interpretable Model-agnostic Explanations) is used. It gives specific reasons for each prediction, helping to see which features affect the outcome. This makes the AI system more transparent and trustworthy. The approach offers a dependable way to detect nitrogen deficiency early, helping farmers take action in time to boost crop health and yield, as shown in the model’s results and visuals.